Application of radio frequency energy in processing of fruit and vegetable products.

Thermal processing is commonly employed to ensure the quality and extend the shelf-life of fruits and vegetables. Radio frequency (RF) heating has been used as a promising alternative treatment to replace conventional thermal processing methods with advantages of rapid, volumetric, and deep penetration heating characteristics. This article provides comprehensive information regarding RF heating uniformity and applications in processing of fruit and vegetable products, including disinfestation, blanching, drying, and pasteurization. The dielectric properties of fruits and vegetables and their products have also been summarized. In addition, recommendations for future research on RF heating are proposed to enhance practical applications for fruits and vegetables processing in future.

The Efficacy of Scouting Technology in Pedicle Screw Placement in the Patients With Degenerative Scoliosis.

Objective: To investigate the curative effect of ball tip technology in pedicle screw placement in the patients with degenerative scoliosis (DS), as compared to traditional freehand technique. Methods: A total of 90 patients with degenerative scoliosis who were admitted to Affiliated Hospital of Hebei Engineering University from October 2019 to October 2021 were selected as the objects in this prospective study. They were randomly divided into an experimental group and a control group with 45 cases in each. The clinical indications, the accuracy of pedicle screw placement, the occurrence of surgical complications, the measurement of spinal and pelvic parameters, the recovery of spinal function and pain degree were recorded and compared within the two groups. Results: After treatment, the operation time, intraoperative blood loss, total number of screws, and time of screwing were compared between the two groups, and the difference was not significant (P > .05). However, the bedding time and the hospital stay were shorter in the experimental group than the control group with difference (P < .05). There was no significant difference in clinical standards and poor implantation in the Gertzbein-Robbins A-E classification between the two groups (P > .05). While the number of perfect placement of screws in the experimental group was higher (P < .05). Before treatment, the Cobbs angle and pelvic incidence-lumbar lordosis (PI-LL) levels of the two groups were comparable (P > .05); after treatment, the Cobbs angle and PI-LL levels of the two groups were lower than those before treatment, and the difference was significant (P < .05). There was no significant difference in Cobbs angle and PI-LL levels between groups (P > .05). Before treatment, the JOA and DOI scores of the two groups were comparable (P > .05); after treatment, the JOA and DOI scores of the two groups were improved (P < .05); the improvement of JOA and DOI scores of the experimental group were better than those in the control group (P < .05). Before treatment, there was no significant difference in the pain degree between the two groups (P > .05); after treatment, the pain of the two groups was improved compared with that before treatment, and the pain degree of the experimental group was lower than that of the control group (P < .05). The incidence of postoperative complications in the experimental group was lower than that in the control group, but there was no significant difference in the total incidence of postoperative complications between the two groups (P > .05). Conclusion: The scouting technique-assisted screw placement can effectively improve the spinal function of patients with degenerative scoliosis, with obvious curative effect and high safety.

Multitarget-Tracking Method Based on the Fusion of Millimeter-Wave Radar and LiDAR Sensor Information for Autonomous Vehicles.

Multitarget tracking based on multisensor fusion perception is one of the key technologies to realize the intelligent driving of automobiles and has become a research hotspot in the field of intelligent driving. However, most current autonomous-vehicle target-tracking methods based on the fusion of millimeter-wave radar and lidar information struggle to guarantee accuracy and reliability in the measured data, and cannot effectively solve the multitarget-tracking problem in complex scenes. In view of this, based on the distributed multisensor multitarget tracking (DMMT) system, this paper proposes a multitarget-tracking method for autonomous vehicles that comprehensively considers key technologies such as target tracking, sensor registration, track association, and data fusion based on millimeter-wave radar and lidar. First, a single-sensor multitarget-tracking method suitable for millimeter-wave radar and lidar is proposed to form the respective target tracks; second, the Kalman filter temporal registration method and the residual bias estimation spatial registration method are used to realize the temporal and spatial registration of millimeter-wave radar and lidar data; third, use the sequential m-best method based on the new target density to find the track the correlation of different sensors; and finally, the IF heterogeneous sensor fusion algorithm is used to optimally combine the track information provided by millimeter-wave radar and lidar, and finally form a stable and high-precision global track. In order to verify the proposed method, a multitarget-tracking simulation verification in a high-speed scene is carried out. The results show that the multitarget-tracking method proposed in this paper can realize the track tracking of multiple target vehicles in high-speed driving scenarios. Compared with a single-radar tracker, the position, velocity, size, and direction estimation errors of the track fusion tracker are reduced by 85.5%, 64.6%, 75.3%, and 9.5% respectively, and the average value of GOSPA indicators is reduced by 19.8%; more accurate target state information can be obtained than a single-radar tracker.

Inorganic-Organic Hybrid Phototransistor Array with Enhanced Photogating Effect for Dynamic Near-Infrared Light Sensing and Image Preprocessing.

Narrow-band-gap organic semiconductors have emerged as appealing near-infrared (NIR) sensing materials by virtue of their unique optoelectronic properties. However, their limited carrier mobility impedes the implementation of large-area, dynamic NIR sensor arrays. In this work, high-performance inorganic-organic hybrid phototransistor arrays are achieved for NIR sensing, by taking advantage of the high electron mobility of In2O3 and the strong NIR absorption of a BTPV-4F:PTB7-Th bulk heterojunction (BHJ) with an enhanced photogating effect. As a result, the hybrid phototransistors reach a high responsivity of 1393.0 A W-1, a high specific detectivity of 4.8 × 1012 jones, and a fast response of 0.72 ms to NIR light (900 nm). Meanwhile, an integrated 16 × 16 phototransistor array with a one-transistor-one-phototransistor (1T1PT) architecture is achieved. On the basis of the enhanced photogating effect, the phototransistor array can not only achieve real-time, dynamic NIR light mapping but also implement image preprocessing, which is promising for advanced NIR image sensors.

Highly Linear and Symmetric Synaptic Memtransistors Based on Polarization Switching in Two-Dimensional Ferroelectric Semiconductors.

Brain-inspired neuromorphic computing hardware based on artificial synapses offers efficient solutions to perform computational tasks. However, the nonlinearity and asymmetry of synaptic weight updates in reported artificial synapses have impeded achieving high accuracy in neural networks. Here, this work develops a synaptic memtransistor based on polarization switching in a two-dimensional (2D) ferroelectric semiconductor (FES) of α-In2 Se3 for neuromorphic computing. The α-In2 Se3 memtransistor exhibits outstanding synaptic characteristics, including near-ideal linearity and symmetry and a large number of programmable conductance states, by taking the advantages of both memtransistor configuration and electrically configurable polarization states in the FES channel. As a result, the α-In2 Se3 memtransistor-type synapse reaches high accuracy of 97.76% for digit patterns recognition task in simulated artificial neural networks. This work opens new opportunities for using multiterminal FES memtransistors in advanced neuromorphic electronics.

Laser-induced dynamic alignment and nonlinear-like optical transmission in liquid suspensions of 2D atomically thin nanomaterials.

Nonlinear optical property of atomically thin materials suspended in liquid has attracted a lot of attention recently due to the rapid development of liquid exfoliation methods. Here we report laser-induced dynamic orientational alignment and nonlinear-like optical response of the suspensions as a result of their intrinsic anisotropic properties and thermal convection of solvents. Graphene and graphene oxide suspensions are used as examples, and the transition to ordered states from initial optically isotropic suspensions is revealed by birefringence imaging. Computational fluid dynamics is performed to simulate the velocity evolution of convection flow and understand alignment-induced birefringence patterns. The optical transmission of these suspensions exhibits nonlinear-like saturable or reverse saturable absorptions in Z-scan measurements with both nanosecond and continuous-wave lasers. Our findings not only demonstrate a non-contact controlling of macroscopic orientation and collective optical properties of nanomaterial suspensions by laser but also pave the way for further explorations of optical properties and novel device applications of low-dimensional nanomaterials.

A redox reversible endoplasmic reticulum-targeted fluorescent probe for revealing the redox status of living cells.

The endoplasmic reticulum (ER) is one of the most important organelles in cells and is involved in protein synthesis, folding and orderly transport. Redox balance is the key to its normal function. In this work, we designed and synthesized an endoplasmic reticulum-targeted fluorescent probe N-Se with selenomorpholine as the redox reversible detection moiety. N-Se could selectively respond to ClO- within only 8 s with a LOD of 28.8 nM. Furthermore, such a response is reversible in the regulation of GSH. Confocal fluorescence imaging confirmed the excellent endoplasmic reticulum targeting ability of N-Se. Thus, it could real-time monitor the dynamic changes of the redox status in the endoplasmic reticulum through the variation of the fluorescence intensity.

A turn-on homodimer fluorescent probe based on homo-FRET for the sensing of biothiols in lysosome: a trial of a new turn-on strategy.

Fluorescence resonance energy transfer (FRET) is often applied to construct fluorescent probes for acquiring high selectivity and sensitivity. According to the FRET theory, a homodimer composed of two identical fluorophores with a small Stokes shift has only weak fluorescence due to homo-FRET between fluorophores, and the fluorescence could be recovered after the destruction of the homodimer. In this study, we designed and synthesized a homodimer fluorescent probe, namely 1,3,5,7-tetramethyl-8-(4'-phenylthiophenol)-boron difluoride-dipyrrole methane dimer (D-TMSPB), based on this turn-on strategy. In D-TMSPB, the disulfide moiety was selected as the response moiety of biothiols, and BODIPY fluorophore was chosen as both donor and acceptor in FRET due to the ultra-small Stokes shifts and obvious overlap of its excitation/emission peak. D-TMSPB exhibited only weak fluorescence. After selective reaction with biothiols, FRET was destroyed and the derivative exhibited strong fluorescence at 514 nm with the limit of detection of about 0.15 µM for GSH. Notably, the derivative of biothiols shows remarkable fluorescence only in acidic conditions, which accords with the internal environment of lysosome. Thus, D-TMSPB was applied to image the biothiols of lysosome in living cells. The turn-on fluorescence of D-TMSPB indicated that homo-FRET is a practical strategy to design turn-on fluorescent probes, particularly for the sensing mechanism based on leaving groups.

Amelioration of human peritoneal mesothelial cell co-culture-evoked malignant potential of ovarian cancer cells by acacetin involves LPA release-activated RAGE-PI3K/AKT signaling.

BACKGROUND: Ovarian cancer is a devastating gynecological malignancy and frequently presents as an advanced carcinoma with disseminated peritoneum metastasis. Acacetin exerts anti-cancerous effects in several carcinomas. Here, we sought to investigate acacetin function in ovarian cancer malignancy triggered by peritoneal mesothelial cells. METHODS: Peritoneal mesothelial cells were treated with acacetin, and then the conditioned medium was collected to treat ovarian cancer cells. Then, cell proliferation was analyzed by MTT assay. Transwell analysis was conducted to evaluate cell invasion. Protein expression was determined by western blotting. ELISA and qRT-PCR were applied to analyze inflammatory cytokine levels. The underlying mechanism was also explored. RESULTS: Acacetin suppressed cell proliferation and invasion, but enhanced cell apoptosis. Furthermore, mesothelial cell-evoked malignant characteristics were inhibited when mesothelial cells were pre-treated with acacetin via restraining cell proliferation and invasion, concomitant with decreases in proliferation-related PCNA, MMP-2 and MMP-9 levels. Simultaneously, acacetin reduced mesothelial cell-induced transcripts and production of pro-inflammatory cytokine IL-6 and IL-8 in ovarian cancer cells. Mechanically, acacetin decreased lysophosphatidic acid (LPA) release from mesothelial cells, and subsequent activation of receptor for advanced glycation end-products (RAGE)-PI3K/AKT signaling in ovarian cancer cells. Notably, exogenous LPA restored the above pathway, and offset the efficacy of acacetin against mesothelial cell-evoked malignancy in ovarian cancer cells, including cell proliferation, invasion and inflammatory cytokine production. CONCLUSIONS: Acacetin may not only engender direct inhibition of ovarian cancer cell malignancy, but also antagonize mesothelial cell-evoked malignancy by blocking LPA release-activated RAGE-PI3K/AKT signaling. Thus, these findings provide supporting evidence for a promising therapeutic agent against ovarian cancer.

Adverse pregnancy outcomes and Helicobacter pylori infection: A meta-analysis.

INTRODUCTION: Helicobacter pylori infection during pregnancy has some adverse effects, but its effects are still conflicting. This meta-analysis study was performed to assess the relationship between H pylori infection and adverse effects during pregnancy. METHODS: Through a systematic literature search up to August 2020, 31 studies included 16 887 pregnant females at baseline and reported a total of 5852 H pylori infection positive and 8196 H pylori infection negative pregnant females, were found recording relationships between H pylori infection and adverse effects during pregnancy. Odds ratio (OR) with 95% confidence intervals (CIs) was calculated between H pylori infection positive vs H pylori infection negative in adverse effects during pregnancy using the dichotomous methods with a random or fixed-effect model. RESULTS: H pylori infection positive during pregnancy was significantly related to higher rate of preeclampsia (OR, 2.68; 95% CI, 2.02-3.56, P < .001), foetal growth restriction (OR, 1.45; 95% CI, 1.26-1.66, P < .001), gestational diabetes mellitus (OR, 2.63; 95% CI, 1.51-4.59, P < .001), and hyperemesis gravidarum (OR, 14.45; 95% CI, 10.24-20.38, P < .001) compared with H pylori infection negative. However, H pylori infection positive during pregnancy was not significantly correlated with spontaneous onset of labour (OR, 1.00; 95% CI, 0.83-1.21, P = .98) compared with H pylori infection negative. CONCLUSIONS: H pylori infection may have an independent relationship with certain adverse effects during pregnancy. H pylori infection positive during pregnancy was significantly related to a higher rate of preeclampsia, foetal growth restriction, gestational diabetes mellitus, and hyperemesis gravidarum compared with H pylori infection negative. This relationship encouraged us to recommend screening and treating females for H pylori infection before and during pregnancy to avoid any possible complications.

Size-controlled synthesis of metal-organic frameworks and their performance as fluorescence sensors.

Metal-Organic Frameworks (MOFs) are of bright promise as new fluorescence sensors because of their accurate framework structure and unique fluorescence properties. Many MOFs have been reported as fluorescence sensors, including bulk-MOF-crystals and nano-MOF-powder. Obviously, the sensing performance of these MOF sensors should be diverse due to their different sizes. However, bulk-MOF-crystals and nano-MOF-powder have completely different dispersibility in solvents, and the effects of this difference on the analytical performance like precision and sensitivity are significant but have not been discussed systematically. To investigate such effects, rodlike bulk-MOFs and nano-MOFs with the same structure but different sizes are required. In this work, we obtained MOFs with a crystal width ranging from 9.7 µm to 170 nm by controlled synthesis, and then proved that they have the same structure by PXRD, SEM, TGA and FTIR analysis. After that, taking folic acid as the target molecule, fluorescent sensing experiments were carried out to compare the sensing performance between bulk-MOFs and nano-MOFs. From the results, we found that nano-MOFs have obviously better dispersity, a lower precipitation speed, a smaller standard deviation, ten times higher fluorescence intensities and a much lower LOD than bulk-MOFs. Finally, we draw a conclusion that nano-MOFs are more in line with the requirements of analytical performance as fluorescence sensors, and the size of MOFs as fluorescence sensors should be as small as possible.

Effects of different manganese precursors as promoters on catalytic performance of CuO-MnOx/TiO2 catalysts for NO removal by CO.

Two different precursors, manganese nitrate (MN) and manganese acetate (MA), were employed to prepare two series of catalysts, i.e., xCuyMn(N)/TiO2 and xCuyMn(A)/TiO2, by a co-impregnation method. The catalysts were characterized by XRD, LRS, CO-TPR, XPS and EPR spectroscopy. The results suggest that: (1) both xCuyMn(N)/TiO2 and xCuyMn(A)/TiO2 catalysts exhibit much higher catalytic activities than an unmodified Cu/TiO2 catalyst in the NO + CO reaction. Furthermore, the activities of catalysts modified with the same amount of manganese are closely dependent on manganese precursors. (2) The enhancement of activities for Mn-modified catalysts should be attributed to the formation of the surface synergetic oxygen vacancy (SSOV) Cu(+)-□-Mn(y+) in the reaction process. Moreover, since the formation of the SSOV (Cu(+)-□-Mn(3+)) in the xCuyMn(N)/TiO2 catalyst is easier than that (Cu(+)-□-Mn(2+)) in the xCuyMn(A)/TiO2 catalyst, the activity of the xCuyMn(N)/TiO2 catalyst is higher than that of the xCuyMn(A)/TiO2 catalyst. This conclusion is well supported by the XPS and EPR results.

MAFE-Net: retinal vessel segmentation based on a multiple attention-guided fusion mechanism and ensemble learning network.

The precise and automatic recognition of retinal vessels is of utmost importance in the prevention, diagnosis and assessment of certain eye diseases, yet it brings a nontrivial uncertainty for this challenging detection mission due to the presence of intricate factors, such as uneven and indistinct curvilinear shapes, unpredictable pathological deformations, and non-uniform contrast. Therefore, we propose a unique and practical approach based on a multiple attention-guided fusion mechanism and ensemble learning network (MAFE-Net) for retinal vessel segmentation. In conventional UNet-based models, long-distance dependencies are explicitly modeled, which may cause partial scene information loss. To compensate for the deficiency, various blood vessel features can be extracted from retinal images by using an attention-guided fusion module. In the skip connection part, a unique spatial attention module is applied to remove redundant and irrelevant information; this structure helps to better integrate low-level and high-level features. The final step involves a DropOut layer that removes some neurons randomly to prevent overfitting and improve generalization. Moreover, an ensemble learning framework is designed to detect retinal vessels by combining different deep learning models. To demonstrate the effectiveness of the proposed model, experimental results were verified in public datasets STARE, DRIVE, and CHASEDB1, which achieved F1 scores of 0.842, 0.825, and 0.814, and Accuracy values of 0.975, 0.969, and 0.975, respectively. Compared with eight state-of-the-art models, the designed model produces satisfactory results both visually and quantitatively.

Abnormal functional connectivity of white-matter networks and gray-white matter functional networks in patients with NMOSD.

Cognitive impairment (CI) has been reported in 29-70% of patients with neuromyelitis optica spectrum disorder (NMOSD). Abnormal white matter (WM) functional networks that correlate with cognitive functions have not been studied well in patients with NMOSD. The aim of the current study was to investigate functional connectivity (FC), spontaneous activity, and functional covariance connectivity (FCC) abnormalities of WM functional networks in patients with NMOSD and their correlation with cognitive performance. Twenty-four patients with NMOSD and 24 healthy controls (HCs) were included in the study. Participants underwent brain resting-state functional magnetic resonance imaging (fMRI) and the Montreal Cognitive Assessment (MoCA). Eight WM networks and nine gray matter (GM) networks were created. In patients, WM networks, including WM1-4, WM1-8, WM2-6, WM2-7, WM2-8, WM4-8, WM5-8 showed reduced FC (P < 0.05). All WM networks except WM1 showed decreased spontaneous activity (P < 0.05). The major GM networks demonstrated increased/decreased FC (P < 0.05), whereas GM7-WM7, GM8-WM4, GM8-WM6 and GM8-WM8 displayed decreased FC (P < 0.05). The MoCA results showed that two-thirds (16/24) of the patients had CI. FC and FCC in WM networks were correlated negatively with the MoCA scores (P < 0.05). WM functional networks are multi-layered. Abnormal FC of WM functional networks and GM functional networks may be responsible for CI.

Wearable and Multiplexed Biosensors based on Oxide Field-Effect Transistors.

Wearable sensors designed for continuous, non-invasive monitoring of physicochemical signals are important for portable healthcare. Oxide field-effect transistor (FET)-type biosensors provide high sensitivity and scalability. However, they face challenges in mechanical flexibility, multiplexed sensing of different modules, and the absence of integrated on-site signal processing and wireless transmission functionalities for wearable sensing. In this work, a fully integrated wearable oxide FET-based biosensor array is developed to facilitate the multiplexed and simultaneous measurement of ion concentrations (H+, Na+, K+) and temperature. The FET-sensor array is achieved by utilizing a solution-processed ultrathin (≈6 nm thick) In2O3 active channel layer, exhibiting high compatibility with standard semiconductor technology, good mechanical flexibility, high uniformity, and low operational voltage of 0.005 V. This work provides an effective method to enable oxide FET-based biosensors for the fusion of multiplexed physicochemical information and wearable health monitoring applications.

An Ion-Mediated Spiking Chemical Neuron based on Mott Memristor.

Artificial spiking neurons capable of interpreting ionic information into electrical spikes are critical to mimic biological signaling systems. Mott memristors are attractive for constructing artificial spiking neurons due to their simple structure, low energy consumption, and rich neural dynamics. However, challenges remain in achieving ion-mediated spiking and biohybrid-interfacing in Mott neurons. Here, a biomimetic spiking chemical neuron (SCN) utilizing an NbOx Mott memristor and oxide field-effect transistor-type chemical sensor is introduced. The SCN exhibits both excitation and inhibition spiking behaviors toward ionic concentrations akin to biological neural systems. It demonstrates spiking responses across physiological and pathological Na+ concentrations (1-200 × 10-3 m). The Na+-mediated SCN enables both frequency encoding and time-to-first-spike coding schemes, illustrating the rich neural dynamics of Mott neuron. In addition, the SCN interfaced with L929 cells facilitates real-time modulation of ion-mediated spiking under both normal and salty cellular microenvironments.

An artificial visual neuron with multiplexed rate and time-to-first-spike coding.

Human visual neurons rely on event-driven, energy-efficient spikes for communication, while silicon image sensors do not. The energy-budget mismatch between biological systems and machine vision technology has inspired the development of artificial visual neurons for use in spiking neural network (SNN). However, the lack of multiplexed data coding schemes reduces the ability of artificial visual neurons in SNN to emulate the visual perception ability of biological systems. Here, we present an artificial visual spiking neuron that enables rate and temporal fusion (RTF) coding of external visual information. The artificial neuron can code visual information at different spiking frequencies (rate coding) and enables precise and energy-efficient time-to-first-spike (TTFS) coding. This multiplexed sensory coding scheme could improve the computing capability and efficacy of artificial visual neurons. A hardware-based SNN with the RTF coding scheme exhibits good consistency with real-world ground truth data and achieves highly accurate steering and speed predictions for self-driving vehicles in complex conditions. The multiplexed RTF coding scheme demonstrates the feasibility of developing highly efficient spike-based neuromorphic hardware.

Functional Changes of White Matter Are Related to Human Pain Sensitivity during Sustained Nociception.

Pain is considered an unpleasant perceptual experience associated with actual or potential somatic and visceral harm. Human subjects have different sensitivity to painful stimulation, which may be related to different painful response pattern. Excellent studies using functional magnetic resonance imaging (fMRI) have found the effect of the functional organization of white matter (WM) on the descending pain modulatory system, which suggests that WM function is feasible during pain modulation. In this study, 26 pain sensitive (PS) subjects and 27 pain insensitive (PIS) subjects were recruited based on cold pressor test. Then, all subjects underwent the cold bottle test (CBT) in normal (26 degrees temperature stimulating) and cold (8 degrees temperature stimulating) conditions during fMRI scan, respectively. WM functional networks were obtained using K-means clustering, and the functional connectivity (FC) was assessed among WM networks, as well as gray matter (GM)-WM networks. Through repeated measures ANOVA, decreased FC was observed between the GM-cerebellum network and the WM-superior temporal network, as well as the WM-sensorimotor network in the PS group under the cold condition, while this difference was not found in PIS group. Importantly, the changed FC was positively correlated with the state and trait anxiety scores, respectively. This study highlighted that the WM functional network might play an integral part in pain processing, and an altered FC may be related to the descending pain modulatory system.

Realizing nearly-zero dark current and ultrahigh signal-to-noise ratio perovskite X-ray detector and image array by dark-current-shunting strategy.

Although perovskite X-ray detectors have revealed promising properties, their dark currents are usually hundreds of times larger than the practical requirements. Here, we report a detector architecture with a unique shunting electrode working as a blanking unit to suppress dark current, and it theoretically can be reduced to zero. We experimentally fabricate the dark-current-shunting X-ray detector, which exhibits a record-low dark current of 51.1 fA at 5 V mm-1, a detection limit of 7.84 nGyair s-1, and a sensitivity of 1.3 × 104 µC Gyair-1 cm-2. The signal-to-noise ratio of our polycrystalline perovskite-based detector is even outperforming many previously reported state-of-the-art single crystal-based X-ray detectors by serval orders of magnitude. Finally, the proof-of-concept X-ray imaging of a 64 × 64 pixels dark-current-shunting detector array is successfully demonstrated. This work provides a device strategy to fundamentally reduce dark current and enhance the signal-to-noise ratio of X-ray detectors and photodetectors in general.

Bidirectional Synaptic Phototransistor Based on Two-Dimensional Ferroelectric Semiconductor for Mixed Color Pattern Recognition.

Optoelectronic synaptic devices capable of processing multiwavelength inputs are critical for neuromorphic vision hardware, which remains an important challenge. Here, we develop a bidirectional synaptic phototransistor based on a two-dimensional ferroelectric semiconductor of α-In2Se3, which exhibits bidirectional potentiated and depressed synaptic weight update under optical pulse stimulation. Importantly, the bidirectional optoelectronic synaptic behavior can be extended to multiwavelengths (blue, green, and red light), which could be used for color recognition. The mechanism underlying the bidirectional synaptic characteristics is attributed to the gate-configurable barrier heights as revealed by the Kelvin probe force microscopy measurement. The α-In2Se3 device exhibits versatile synaptic plasticity such as paired-pulse facilitation, short- and long-term potentiation, and long-term depression. The bidirectional optoelectronic synaptic weight updates under multiwavelength inputs enable a high accuracy of 97% for mixed color pattern recognition.